Richard Lehman’s journal review—21 September 2015

richard_lehmanNEJM 17 Sep 2015 Vol 373
1095 Well, here’s a paper that nearly caused me to stop breathing. It was certainly followed by a sharp intake of breath. Cheyne-Stokes breathing is common in advanced heart failure, and so is central sleep apnoea, which causes frequent periods of nocturnal hypoxia in these patients. Thus continuous positive airways pressure should help them feel better and live longer. This trial shows that it actually causes them to die faster. And, unlike in a similar trial which was reported recently, it did not make them feel better. This is important, because most people with heart failure would prefer feeling better to living longer. So we now know that for men aged around 70 with marked reductions in ejection fraction, CPAP is effective in reducing sleep-disordered breathing, but is unlikely to have any symptomatic benefit and increases the risk of cardiovascular death. This should not preclude further trials in different people with other types of heart failure, but it probably will.

1106 Amyloidosis. In 35 years of British general practice, I can’t remember coming across a case. However, this is the NEJM and we expect it to sing of such lofty themes: “The amyloid fibril deposits that cause systemic amyloidosis always contain the nonfibrillar normal plasma protein, serum amyloid P component (SAP). The drug (R)-1-[6-[(R)-2-carboxy-pyrrolidin-1-yl]-6-oxo-hexanoyl]pyrrolidine-2-carboxylic acid (CPHPC) efficiently depletes SAP from the plasma but leaves some SAP in amyloid deposits that can be specifically targeted by therapeutic IgG anti-SAP antibodies.” GlaxoSmithKline has actually made such antibodies and deployed them without harm in 15 patients with systemic amyloidosis in a phase 1 trial. Using sophisticated imaging, they showed that amyloid deposits in the liver and kidneys shrank down over a six week period. Here’s a space worth watching if you know anyone with amyloid, but note that no patients with cardiac amyloid (the most dangerous type) were included.

1125 “Percutaneous Implantation of an Entirely Intracardiac Leadless Pacemaker.” How cool is that? “The leadless cardiac pacemaker met prespecified pacing and sensing requirements in the large majority of patients. Device-related serious adverse events occurred in approximately 1 in 15 patients. (Funded by St. Jude Medical)” So not very cool, really. If you went into heart block tomorrow, would you want one of these?

OL We seem to have a new class of drugs which reduce cardiovascular death, heart failure, and all cause mortality by over 30% in high risk individuals. Well, one of the class might anyway. It is a sodium-glucose cotransporter 2 inhibitor called empagliflozin. Come on, say that name aloud: you know you can do it. On the basis of the EMPA-REG OUTCOME trial, the gliflozins could one day (and many trials hence) become as important as the statins. To know how the participants were selected and discover their baseline characteristics, you have to go to another paper in a rather obscure journal.
All 7020 patients had known coronary, cerebrovascular, or peripheral arterial disease, with a mean age of 63 and mean BMI 30.6. In typical industry sponsored fashion, they were recruited from 590 sites in 42 countries. A quarter of those given empagliflozin abandoned the drug before completion of the trial, but in those who were randomised to take either 10mg or 25mg daily, death from cardiovascular causes was 3.7%, vs. 5.9% in the placebo group; (38% relative risk reduction), hospitalisation for heart failure was 2.7% and 4.1%, respectively (35% relative risk reduction), and death from any cause 5.7% and 8.3%, respectively (32% relative risk reduction). Oh, and by the way, all these individuals also had a glycated haemoglobin of 7-9%. But we know that HbA1c in itself could have made only a slight contribution to their overall risk, as shown in a new analysis by Jamie Jarmul and colleagues. Given the effect size in these high risk individuals, it would be wrong for future trials of sodium-glucose cotransporter 2 inhibitors to be limited to people with so called type 2 diabetes, simply because these drugs make people pee out a bit of sugar and reduce HbA1c by 0.6% or so. It is just possible, however, that the trial demonstrates the harm of sulfonylureas and insulin given disproportionately to the placebo group, rather than showing a real benefit from empagliflozin. We need new trials in different risk groups, and also the full data from this one to guide decision making in the meantime.

JAMA 15 Sep 2015 Vol 314
1142 I think I got spectacles when I was about 7. Suddenly the world became much sharper and smaller, and for a day or two I tripped over things. My early years were lived a minute’s run from the wide open West Common in Lincoln, where I spent most of the summer days. From this narrative you will gather that I am not Chinese, I did not lack time outdoors, that I was discovered to be short sighted, and have worn glasses since childhood. I realise that these facts of themselves are of little global importance, but they do allow me to introduce you to a randomised trial, which was aimed at reducing the alarming increase in myopia among Chinese schoolchildren. “Among 6 year old children in Guangzhou, China, the addition of 40 minutes of outdoor activity at school compared with usual activity resulted in a reduced incidence rate of myopia over the next 3 years.” Clearly didn’t work for me.

1159 Next we’re off to France, where a large registry based study finds that after adjusting for age, sex, and co-morbidities, people have worse outcomes after hip fracture surgery than after elective hip surgery. Well yes. People who fall over and break their hips are not the same as people who just have hip arthritis, adjust as you may. Nor is it possible to follow the same surgical routines.

JAMA Intern Med Sep 2015
OL Just to the south of central Rome, next to the Tiber, there is a huge protuberance called Monte Testaccio. It consists entirely of broken pottery from the mid to late imperial period, and Wikipedia tells us that “it has been estimated that the hill contains the remains of as many as 53 million olive oil amphorae, in which some 6 billion litres of oil were imported.” Most of this came from Spain, which remains the world’s largest exporter of olive oil. Olive oil was also one of the interventions in the Spanish PREDIMED trial, to which 4282 women aged 60 to 80 years and at high cardiovascular disease risk were recruited between 2003 and 2009. Participants were randomly allocated to a Mediterranean diet supplemented with extra virgin olive oil, a Mediterranean diet supplemented with mixed nuts, or a control diet (advice to reduce dietary fat). One of the prespecified secondary outcomes was the occurrence of new breast cancer, and this open access report shows that the olive oil group fared best of the three. The result is statistically significant, but since there were only 35 cases in the whole trial population over less than five years, it seems a bit early to conclude that olive oil prevents breast cancer.

Ann Intern Med 15 Sep 2015 Vol 163
437 I keep having a dig at the arbitrary definition of “type 2 diabetes,” but we can all agree that there is a common syndrome in which blood glucose levels rise inexorably over time due to a combination of insulin resistance and beta-cell exhaustion, right? And that this carries an increased risk of various bad vascular outcomes, mostly cardiac but also peripheral, ophthalmic, and renal? And that the solution must lie in prevention by promoting a better diet and more exercise? This seems to be borne out by a new American systematic review of Combined Diet and Physical Activity Promotion Programs to Prevent Type 2 Diabetes Among Persons at Increased Risk. “Conclusion: Combined diet and physical activity promotion programs are effective at decreasing diabetes incidence and improving cardiometabolic risk factors in persons at increased risk. More intensive programs are more effective.” But effective at what? The hidden party pooper here is that if you go to the section on cardiovascular outcomes, you will see that there is no evidence of any reduction in real events as opposed to surrogates like sugar and “cardiometabolic risk factors.” And that is truly disappointing. For more about this, read the magnificently intelligent deconstruction of the NHS England’s Diabetes Prevention Programme in The BMJ.

Lancet 19 Sep 2015 Vol 386
OL Big surveys of the health effects of alcohol keep appearing, and this one justifies its existence by claiming to be the first to survey self-reported alcohol consumption and health outcomes over a range of countries (12 in all) at different levels of economic development. Overall, mortality in the median follow-up period of 4.3 years was higher in heavy drinkers than for non-drinkers, with a hazard ratio of 1.31 (CI 1.04-1.66). However, high intake was associated with less cardiovascular disease and did not affect the risk of stroke. Those of us who like to finish the day with a bottle of wine can avoid injury by staying at home. We take our chances with a higher risk of some cancers. Liver disease, surprisingly, does not figure in this survey.

The BMJ 19 Sep 2015 Vol 351
Back in November 2011, I set up a Google group called PATH (publish all trials on humans) and wrote an editorial in The BMJ on the subject with Elizabeth Loder. In January 2013, Sense About Science and Ben Goldacre launched AllTrials with enormous success. It seems that another group was also set up in 2013, called RIAT, “restoring invisible and abandoned trials.” My initial group was just a talking shop; AllTrials is an international public campaign; and now RIAT has gone and published a reanalysis of one of the trials which serves as an example for all of us. They deserve congratulations, as does GlaxoSmithKline for releasing the dataset and itself joining AllTrials. It is no secret that Study 329 was an example of all that is bad and harmful in industry reported trials, and that paroxetine, far from being a suitable antidepressant for use in adolescents, is ineffective and potentially dangerous. GSK is mercifully no longer the same organisation as SmithKline Beecham was in the 1990s. And we must hope that the 12 North American academic psychiatry centres, which were paid to recruit and lend their names to this study, have since changed their ways too. But the price of freedom from fraud is eternal vigilance.
Peter Doshi meditates on the lack of apology, retraction, or disciplinary action in a separate feature piece
and there is also a separate editorial on liberating data from clinical trials. But I will let Fiona Godlee have the last word: “The RIAT re-analysis marks a new chapter in the story of Study 329, showing the remarkable power of open data. But it also shows how much our current systems are failing patients and the public. It should not have taken 14 years to get to this point. It shows that we need regulation, and perhaps legislation, to ensure that the results of all clinical trials are made publicly available and that individual patient data are available for legitimate independent third party scrutiny.”

Fungus of the Week: Calvatia gigantea

According to Wikipedia, giant puffballs can grow to a weight of 20 kilograms. Even an average one weighing a pound or two, found in perfect condition, can take quite a bit of eating. They grow in the same place each year and it is said that in some country districts, annual puffball feasts would be held to honour their appearance. They can occasionally be found in enormous fairy rings on old undisturbed pasture land.

Sliced pure white puffball fried in bacon fat and lightly sprinkled with finely chopped parsley and chives is a dish for a king. If found at the right stage, puffballs like this will keep in the refrigerator for many days. But beware the gradual onset of yellowness, which marks spore formation and the end of edibility.

An average sort of giant puffball contains 20 000 000 000 000 spores. In his wonderful 1953 classic on Mushrooms and Toadstools in the New Naturalist series, John Ramsbottom mused on this:
“This species heads the list for single fruit-bodies and is probably the most prolific organism living on our planet . . . What if the spores all fulfilled their function and produced fruit-bodies? Seven billion* of these of average size placed end to end would put a girdle round about the earth more than five times; if their spores were equally successful the resulting fruit-bodies would stretch twice to the sun and back, and form a mass eight hundred times the weight of the globe. We are on safe ground in asserting this does not happen.”

*in 1953 England, one billion meant one million x one million.